In this Project, we propose to continue the investigation of potentially modifiable nutritional and hormonal risk factors for incident ovarian cancer. This work will use prospectively collected questionnaire data, biological samples, and paraffin-embedded ovarian tumor tissue from women in the Nurses' Health Study (NHS) and NHSII. Although several hypotheses regarding ovarian cancer etiology have been proffered, few modifiable lifestyle risk factors have been identified. We will examine the possible preventive role of vitamin D in ovarian cancer, with the goal of identifying subpopulations who could most benefit from intervention. We also will evaluate three dietary factors: flavonoids, which may reduce risk; lactose in adolescence and adulthood; and acrylamide intake and red blood cell adduct levels, both of which may increase risk. We also will evaluate the role of melatonin and shift work in ovarian carcinogenesis. In addition, we propose to measure circulating markers of inflammation (CPR, IL-6, and TNFa-R2) and tumor markers associated with inflammation (e.g., STATs). We also propose for the first time to focus specifically on fatal ovarian cancer, with an effort to identify risk factors that may predispose women to either having fast-growing tumors or a peritoneal environment susceptible to metastasis. In this same vein, we plan to evaluate hormone receptors in ovarian tumors and their relation to risk factors, with an emphasis on exploring a new model of ovarian carcinogenesis (slow-growing tumors vs. fast-growing tumors) as a way to classify tumors in epidemiologic studies. The Project shares common biological themes with Projects 1 and 2, including the role of vitamin D, melatonin, and the inflammation pathway and interacts closely with Project 4 for methodologic issues. It also shares with the other Projects a strong administrative and scientific infrastructure provided by Cores A (cohort follow-up and data base maintenance), B (confirmation of cancer and cause of death), C (management of the biospecimens) and D (leadership and data analysis). The findings should enhance understanding of the etiology and progression of ovarian cancer and provide guidance for women and their health care providers in their efforts to reduce suffering and death from this disease.